With all the discussion about where the best city to host the championship is, I thought I’d give another look at my post from a few years ago on driving distances to various US cities.
The map shows the time it would take the average FRC team to drive from their county to the city shown. Now instead of choosing the top 100 cities by population, I chose the cities that have convention centers, according to this list from Wikipedia (excluding Honolulu). The driving time calculation is again done by Project OSRM. Team locations are taken from https://frcmap.com/. Unlike the previous model which included all drivable teams that ever existed, this map only includes the ~2800 teams in the contiguous US, Mexico, and Canada that competed in 2022.
From this analysis, Houston is ranked 28th out of 46 cities, with an average driving time of ~24 hours. The shortest distance cities are Chicago and Indianapolis, with driving distances of 16.4 and 16.5 hours, respectively. You can see a list of all of the cities, as well as the code used to generate the list, here.
This analysis only takes into account driving distance and the availability of a convention center in the city. It doesn’t necessarily mean that the city has all of the resources to host a championship (international airport, hotels, etc). It also doesn’t factor in that after a certain distance teams will prefer to fly than drive. I tried this with my first analysis, but it ended up being too subjective of a metric to really be useful.
II left Einstein at Saturday 5pm and made it home Monday at 9am. The nights were for sleeping as much as possible and I stopped often for bathroom breaks and food. It’s a long trip. If FIRST is actually placing an idle gap between week 6 and the championship I will take the vacation days slotted for MSC and use it for the trip back. I can actually stay at hotels too instead sleeping at rest stops in my car.
To be honest, I’m kinda disappointed that Alaska and Canada aren’t included in this.
(/s for Alaska, but Toronto isn’t THAT far from Detroit, so it could be an option if it has a convention center)
Canadian and Mexican teams are included in the calculation (as long as they’re within ~130 miles of a major city, due to an optimization in the code).
If you’re talking about hosting the championship in Canada, that doesn’t really seem feasible at the moment with the vast majority of the teams being from the US. International travel with a school team is significantly more difficult than domestic travel. I know this personally, since we have to travel internationally to the US in order to attend champs. It doesn’t seem like FIRST would want to host their main event somewhere that puts an unnecessary added barrier for most of their teams to attend as long as there are options within the US that don’t have this barrier.
This is cool. I’m not sure that the drive time for the average team necessarily is the most helpful way to visualize this, though.
I don’t know exactly what I want, but I feel like what’s missing in this analysis is that certain locations (like the coasts) are very close to some teams and very far for others, creating funky averages. I suppose my gripe is that this argues for the “fairest” location being somewhere like St. Louis or Omaha, but that fairness might suck for most teams. A 17 hour drive is bad, so if very few teams are close, that’s not good just because it’s fair. I’d advocate for an event which is close to a significant number of teams – not the geographic center – since those further away will fly anyways. Maybe the metric I’m looking for is teams within a drivable distance to each venue, but as you acknowledged, that’s subjective.
Perhaps if the data is easily available, it would be nice to have a table with the convention center cities listed and then the number of teams within 0-10 hours, 11-20 hours, etc.
My thought is that you could omit teams from the average hours driven calculation if the hour count is greater than some number (say 24 hours, because 24 bus rides suck), since those teams will just fly regardless of it being 25 hours of 40 hours away.
That makes a lot of sense since you’re minimizing the distance to Michigan and NE teams. My guess is that you have a lot of <5 hour travel teams and the rest.
If you’re making the assumption that every continental US team drives, I would think that rather than average you would want to compare SD so you would pic the place that doesn’t really serve anyone the best but is the most fair.
That being said, I think champs should be at Disney World again (literally the worst place by the fairness metric I just stated)!
This is very cool, but others have made some good points about the limitations of average driving time as a metric. While maps aren’t great when you have multiple axes you’d like to visualize, I’d love to see a scatter plot of different cities with these axes:
number/percentage of teams within driving distance
average driving time for teams within that driving distance
Just saying, the I-X center in Cleveland is almost twice as large as GRB, and this means I wouldn’t have to take a plane to get to worlds, so Cleveland is clearly the ideal location for worlds (in my opinion)
By this logic, either California, Texas, or Michigan would be the best place. (Texas I would argue against in light of recent anti LGBTQ+ laws, among other things)
I’m OK with champs in Michigan or California.
No you would want to get the lowest SD which would mean that everyone has to travel roughly the same amount. So arguably Kansas (or around there) would be the best place.
Back when the two championship setup was announced, FIRST put together a committee to help decide which regions would go to which Championship. As part of that committee I whipped up the following map, which shows which regions were within an 18 hour drive of either Championship location. Not exactly what you’re looking for, but I figured it might be useful to look at.
I think California would be best, and it should be close to the PNW to reduce driving time there as well, and it shouldn’t be in the Bay Area because that’s too expensive, and it should be a major college campus near a mid sized city, so it should be in Davis, California!
I wouldn’t assume that teams farther away will just fly there. I know of teams from Ontario (Canada) that bussed to Houston. Flights are not necessarily available anyway. That’s a lot of time on the road. probably too much vacation time for many mentors too, and perhaps too much time out of class for some students.
Had to rerun the 8+ hour script to get the raw data, but I was able to do some more analysis this morning based on the suggestions in this thread. I chose to look at three different cutoffs: 10, 15, and 20 hours driving. You can see the raw data and the code used here.
This is essentially the same data as the sets of maps above, just presented together on a graph instead of on separate maps. “Better” on these graphs is closer to the top-left corner (more teams in the range, who have to drive a shorter distance)
Edit to note: Houston is in the bottom (few teams, with a medium length avg drive) for all three cutoffs. The top teams are, unsurprisingly, consistently in the area around southern Michigan (Detroit, Cleveland, Columbus, Grand Rapids, Chicago, etc)
I think these graphs clearly show that almost anywhere in the Midwest would be more accessible to the most number of teams than Houston.
As evidenced by the district model, the program grows as a direct effect of accessibility. So it confounds me that FIRST would do something that removes that accessibility from most of their teams.
